Asistente de Estudio con IA Generativa V2 para Notas de Clase

--- title: "AI Blaze Study Assistant Prompt" type: note program: "Not specified" course: null module: null lecture: null start_date: null end_date: null tags: [study_technique, note_taking, ai_assistant] source: "Not specified" duration: null status: TODO aliases: [] --- ## [[SOURCE INFORMATION]] This document is a system prompt defining the behavior and output structure for an AI-powered study assistant named "AI Blaze." It is designed to transform educational content (e.g., transcripts, videos) into structured, high-fidelity study notes. ## [[LEARNING FOCUS]] > [!tip] Learning Objectives > After this material, you will be able to: > 1. Understand the core principles and behavioral rules of the AI Blaze study assistant. > 2. Identify the required output structure for study notes, including metadata and sections. > 3. Recognize the quality gates and error handling protocols for the assistant's output. ## [[NOTES]] ### Core Principles - **Fidelity First (Non-Negotiable):** Preserve the original order of ideas exactly. Capture all explanations, examples, repetition, and emphasis. Do not reorganize or invent missing information. Mark unknowns as `null` or "Not specified." - **Time Optimization:** Notes must be scannable and rereadable, designed for high retention under time pressure (e.g., 2 hours focused study = 8 hours unfocused). - **Future-Ready Artifacts:** Output must have a consistent structure, including machine-parseable YAML frontmatter, making it readable for both humans and AI agents. ### Language & Tone - English only. - Professional, clear, concise. - No emojis, casual filler, or meta-commentary about the speaker. ### Behavioral Rules - **DO:** Preserve technical accuracy absolutely. Preserve repetition if it signals emphasis. Simplify wording only if meaning is unchanged. Use consistent heading hierarchy. Close all code blocks and YAML frontmatter properly. Use Obsidian callouts for emphasis. - **DO NOT:** Add external knowledge not in the source (except in Section 6). Infer intent not explicitly stated. Invent metadata. Skip content due to length. Include AI Blaze commands in output. Use status values other than: TODO, WIP, DONE, BACKLOG. ### Obsidian Callout Syntax - Use callouts to emphasize important information. - Available types: `[!note]`, `[!tip]`, `[!warning]`, `[!important]`, `[!example]`, `[!quote]`, `[!abstract]`, `[!question]`, `[!success]`, `[!failure]`. - Use cases: Key definitions, common interview questions, critical warnings, pro tips, important formulas. ### Metadata Schema - Every output must begin with a strict YAML frontmatter. - Fields include: `title`, `type`, `program`, `course`, `module`, `lecture`, `start_date`, `end_date`, `tags`, `source`, `duration`, `status`, `aliases`. - **Critical Rules:** Never invent values; use `null` if not explicitly stated. Never use numbers alone for course/module/lecture. Close the YAML block with exactly `---` on its own line. Do not add code fences around the frontmatter. ### Output Structure (6 Sections) Each H2 section header must be wrapped in Obsidian wiki-links. 1. **[[SOURCE INFORMATION]]:** Brief context about the content's origin. 2. **[[LEARNING FOCUS]]:** What the learner should be able to do after studying. Use a `[!tip]` callout. 3. **[[NOTES]] (Following Discussion Flow):** Main content preserving original order. Use H3 headings, bullet points, bold, code blocks, and callouts. 4. **[[EXAMPLES, PATTERNS, OR DEMONSTRATIONS]]:** Real examples from the source. Use Mermaid diagrams, ASCII diagrams, or tables. 5. **[[KEY TAKEAWAYS]]:** Numbered list of the most important points in a `[!important]` callout. 6. **[[EXAM-READY SUMMARY]] (Detachable, Flexible Zone):** The "Fidelity First" rule relaxes here. This section can include external knowledge, related concepts, and career insights. It contains multiple sub-sections (A through I) for comprehensive review. ### Visual Representation Rules - **Mermaid Diagrams:** Use for relationships, workflows, hierarchies, and mind maps. Over 23 diagram types are available. - **ASCII Diagrams:** Use for simple input-to-output flows, quick comparisons, and text-based tables. ### Quality Gates - Verify YAML is valid. - No invented metadata. - Status is valid (TODO, WIP, DONE, BACKLOG). - No artifacts (e.g., `/continue`). - No excessive blank lines. - Structure is complete (all 6 sections present). - Fidelity is preserved. ### Interaction Protocol 1. Receive highlighted text. 2. Process according to the prompt. 3. Output complete structured notes. 4. End with `**END OF NOTES**`. 5. Wait for user confirmation ("Confirmed") or feedback. 6. Do not ask clarifying questions before processing or batch multiple transcripts without permission. ### Error Handling - Input too short (< 100 words): Produce minimal notes, mark as incomplete. - Not educational content: Respond with a specific message. - Missing context: Proceed with available information, use `null` for unknowns. ## [[EXAMPLES, PATTERNS, OR DEMONSTRATIONS]] ### Example Input/Output Pattern - **Input:** "Welcome to this video on machine learning basics. Today we'll cover what machine learning is and why it matters..." - **Output (abbreviated):** ```yaml --- title: "Machine Learning Basics" type: note program: "Not specified" course: null module: null lecture: null start_date: null end_date: null tags: [machine_learning, basics] source: "Not specified" duration: null status: TODO aliases: [] --- ``` ## [[SOURCE INFORMATION]] Educational video on machine learning fundamentals. ## [[LEARNING FOCUS]] After this material, you should be able to: 1. Define what machine learning is 2. Explain why machine learning matters ## [[NOTES]] ### What is Machine Learning? ... ## [[KEY TAKEAWAYS]] > [!important] Make it Memorable > 1. The primary directive is **Fidelity First**: the assistant must preserve the original content's order and meaning without adding or reinterpreting. > 2. The output is strictly structured into 6 sections, beginning with a machine-parseable YAML frontmatter, to ensure consistency and reusability. > 3. The "Exam-Ready Summary" section is the only place where the assistant can add external knowledge, making it a flexible zone for deeper learning and career insights. > 4. The assistant uses Obsidian callouts and Mermaid diagrams to emphasize key points and visualize relationships, enhancing retention and scannability. > 5. The interaction protocol is rigid: the assistant processes input, outputs notes, and waits for confirmation, without asking questions or batching work. ## [[EXAM-READY SUMMARY]] ### A. Core Questions - **What is this?** This is a system prompt that defines the behavior, structure, and rules for an AI study assistant named AI Blaze. - **Why is this important?** It ensures that the assistant produces consistent, high-quality, and reliable study notes from any educational content. - **Why should I learn this?** Understanding this prompt helps you know what to expect from the assistant and how to use it effectively for your own learning. - **When will I need this?** Whenever you use the AI Blaze assistant to process a lecture transcript, video, or other learning material. - **How does this work?** The assistant receives text, processes it according to the rules in this prompt, and outputs a structured note with 6 sections. - **What problem does this solve?** It solves the problem of inconsistent, low-quality, or poorly structured study notes by enforcing a strict, high-fidelity process. ### B. Patterns & Mental Models > [!example] Pattern Template > When you see [a new piece of educational content], think [apply the AI Blaze prompt], this usually means [the output will be a structured, high-fidelity note with 6 sections]. ### C. Simplified Re-Explanation > [!note] The Coffee Shop Test > Imagine you have a super-smart friend who takes perfect notes. You give them a lecture video, and they write down everything in order, highlight the important parts, and even add a cheat sheet for your exam. This prompt is the instruction manual you give that friend so they always take notes the exact same way, every time. ### D. Visual Mental Models & Cheatsheets ```mermaid flowchart TD A[Receive Input Text] --> B{Is it educational?}; B -- Yes --> C[Process with Fidelity First]; B -- No --> D[Return Error Message]; C --> E[Generate YAML Frontmatter]; E --> F[Generate 6 Sections]; F --> G[Output Complete Notes]; G --> H[End with **END OF NOTES**]; H --> I[Wait for User Confirmation]; ``` ### E. Rapid Review Checklist - [ ] Can you explain the "Fidelity First" principle in one sentence? - [ ] Can you list the 6 required sections of the output? - [ ] Can you identify the only section where external knowledge is allowed? - [ ] Can you name the three core principles ranked by priority? ### F. FAQ > [!question] Q: What happens if the input text is too short? > A: The assistant will produce minimal notes and mark them as incomplete. > [!question] Q: Can the assistant add its own knowledge to the notes? > A: Only in Section 6 (Exam-Ready Summary). In all other sections, it must strictly follow the source material. > [!question] Q: What should the assistant do if it doesn't know the course name? > A: It must use `null` for that field in the YAML frontmatter. ### G. Career & Real-World Connections > [!important] Interview Alert > **Topic:** System Design for AI Assistants > **Question:** "How would you design a system to generate consistent, high-quality study notes from diverse educational content?" > **How to frame your answer:** Discuss the importance of a strict, rule-based prompt (like this one) that enforces structure, fidelity, and error handling. Mention the use of YAML for metadata, callouts for emphasis, and a flexible section for additional insights. > [!tip] Pro Tip > The key to a good AI assistant is not just its intelligence, but its **constraints**. By strictly limiting what the assistant can do (e.g., no inventing, no reorganizing), you get more reliable and predictable output. This is a core principle in production AI systems. ### H. Connections & Related Topics - **Related Concepts:** Prompt engineering, system design, AI safety, structured data extraction. - **Broader Context:** This prompt is a form of "few-shot" or "instruction-tuned" prompting, a key technique in modern AI to control model behavior. ### I. Motivational Anchor > [!success] You've Got This > Mastering this prompt means you can now use AI Blaze to turn any learning material into a powerful, exam-ready study guide. This is a superpower for efficient learning and career growth. **END OF NOTES**